Poster De Conférence Année : 2023

An Open-Source Framework for Efficient Numerically-Tailored Computations

Résumé

Many scientific computing problems can be reduced to Matrix-Matrix Multiplications (MMM), making the General Matrix Multiply (GEMM) kernels in the Basic Linear Algebra Subroutine (BLAS) of interest to the high-performance computing community. However, these workloads have a wide range of numerical requirements. Ill-conditioned linear systems require high-precision arithmetic to ensure correct and reproducible results. In contrast, emerging workloads such as deep neural networks, which can have millions up to billions of parameters, have shown resilience to arithmetic tinkering and precision lowering.

Fichier principal
Vignette du fichier
BSC_Symposium_10_Louis_Ledoux_final.pdf (2.81 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04094835 , version 1 (11-05-2023)

Licence

Identifiants

  • HAL Id : hal-04094835 , version 1

Citer

Louis Ledoux, Marc Casas. An Open-Source Framework for Efficient Numerically-Tailored Computations. BSCSymposium23, May 2023, Barcelona (Online), Spain. 10th International BSC Severo Ochoa Doctoral Symposium 2023, 2023. ⟨hal-04094835⟩
81 Consultations
177 Téléchargements

Partager

  • More